We propose a 5 year study of the interactions between reproductive tract infection, inflammation and genital HIV shedding in women. We plan to examine the hypothesis that increased genital tract HIV-1 shedding occurs in association with altered vaginal flora, cervicitis, and endometricitis, all of which are usually sub-clinical conditions that describes a continuum of ascending genital tract infection.. An increase in vaginal and cervical proteolytic enzymes may promote the traffic microbes through the cervix to the endometrium to produce cervicitis and endometritis. The mechanisms by which abnormal vaginal flora, cervicitis increase genital HIV shedding include depletion of protective H2O2 lactobacilli, increased vaginal pH, decreased reduction-oxidation potential and altered vaginal and cervical inflammatory cytokines, which together selectively up-regulate LTR transcription through NF-kappaB. As such, oral antibiotic treatment of BV and cervicitis should decrease HIV shedding. We will examine the following specific aims: 1. To examine the associations between vaginal flora and HIV RNA concentrations in endocervical and vaginal fluid. We hypothesize that increased anaerobes, G. vaginalis and M. hominis in vaginal flora caused increased cervical and vaginal HIV RNA replication and that H2O2- producing Lactobacillus are productive. 2. To study the relationship between reproductive tract inflammation and genital HIV shedding. We hypothesize that an inflammatory response in the vagina, cervix and uterus results in increased genital HIV shedding. 3. To determine whether antibiotic treatment for bacterial vaginosis and cervicitis decreased genital HIV shedding. We hypothesize that oral antibiotic treatment will re-establish normal vaginal flora and decrease local inflammation, resulting in decreased endocervical HIV shedding. Women will be enrolled from 2 sites in the United States. In addition, a BV treatment study will be carried our in parallel in a cohort of HIV- infected Kenyan women not on anti-retroviral therapy. These studies should help to define the inter-relationships between altered vaginal flora, upper genital tract inflammation, the host inflammatory response and genital HIV shedding. By including an African cohort, we will learn about the relative contributions of genital tract infection and inflammation and anti-retroviral therapy to genital HIV load. Finally, we will learn whether antibiotic therapy with the goal to establish normal vaginal flora and decreases cervicitis and endometritis has the potential to decrease HIV shedding in the female genital tract in both anti-retroviral- experienced women and women without access to HIV treatment.